Trends in alternative medicine use in the United States, 1990-1997

Results of a follow-up national survey 

Authors Eisenberg D, Davis R, Ettner S, Appel S, Wilkey S Van Rompay M, Kessler R.
Source JAMA. 280:1569-75. November 11, 1998. 
Institutions Center for Alternative Medicine Research and Education, Beth Israel Hospital, Boston; Department of Health Care Policy, Harvard.
Support National Institutes of Health; several private foundations. 


Alternative medicine (defined here as medical interventions not taught in medical schools and not generally available in US hospitals) is being increasingly used around the world. The extent to which Americans turn to alternative methods was documented by a telephone survey in 1990. The current study was designed to examine the same question in 1997, both to elucidate the current state of affairs and to look at evolving trends.


This study was based on two telephone surveys conducted in the United States. The first survey, conducted in early 1991, was considered representative of 1990 data. The second, conducted between November, 1997 and February, 1998, was considered representative of 1997. The two surveys were similar in structure.

Survey method

The surveys were carried out by random digit telephone dialing and were limited to one English-speaking respondent, 18 years or older, from each household contacted.

The target sample size was 1,500 for the 1990 survey and 2,000 for the 1997 survey. The latter target size was based on a power calculation: an 80% probability of detecting an increase in the overall use of alternative medicine from 34% (noted in 1990) to 39% or more.

Because of an overall trend towards lower response rates to telephone surveys, a financial incentive was offered for participation in 1997 ($20 per participant, and $50 to convert some refusers to participants). No financial incentive had been offered in 1990.

Survey results were weighted to reflect geographic variation in household size and response rate and were also weighted on sociodemographic variables.

Survey questions

Although the 1997 survey was more detailed and lengthy than the 1990 survey, most of the questions posed in 1990 were again posed in 1997. These replicated questions, which enable the comparison of the two surveys, are the focus of this report. Survey questions explored:

  • Major medical problems during the preceeding 12 months and consultations with medical doctors for these problems.

  • Use of 16 prespecified alternative medical therapies during the previous year and lifetime use of these therapies, both supervised by a practitioner and unsupervised.

  • Degree of insurance coverage for each of the therapies for which a practitioner was consulted.

  • Estimation of the cost of alternative therapies, taking into account visits to practitioners, degree of insurance coverage and use of products such as vitamins, herbs and commercial diets.


Overall use in 1997

According to the data, 42% of the population used at least one alternative therapy in 1997. Use was more frequent among women than men (49% vs. 38%), and was most frequent (50%) in the 36-49 year age bracket. Use was higher in those with college education (51%) and with incomes over $50,000 (48%).

Use of alternative therapies

Of the 16 therapies surveyed, the most commonly used in 1997 were:

  • Relaxation techniques (primarily meditation), used by 16.3% of the population (up from 13.1% in 1990). Of those who used this technique, 15.3% consulted a practitioner in 1997.
  • Herbal medicine, used by 12.1% (compared with only 2.5% in 1990).  Of users in 1997, 15.1% consulted a practitioner.
  • Massage, used by 11.1% (compared with 6.9% in 1990). Of 1997 users, 61.6% consulted a practitioner.
  • Chiropractic, used by 11% (10.1% in 1990). Of users, 89.9% consulted a practitioner in 1997.
The remaining 12 therapies and their use in 1997 were: spiritual healing by others (7.0%), mega-vitamins (5.5%), self-help group (4.8%), imagery (4.5%), commercial diet (4.4%), folk remedies (4.2%), lifestyle diet (4.0%), energy healing (3.8%), homeopathy (3.4%), hypnosis (1.2%), biofeedback (1.0%), acupuncture (1.0%).

Overall, 42.1% of the population used one or more of these techniques in 1997, compared with 33.8% in 1990.

Visits to practitioners

Based on the average number of users, percentage of users who consulted a practitioner and average number of visits per user who consulted a practitioner, the authors estimated the numbers of visits to alternative medicine practitioners in 1997. For all therapies combined, the estimated number of visits was 628,825,000 (which was 62% more than the total number of visits to primary care physicians in 1997).

Visits to chiropractors (191,886,000) and to massage practitioners (113,723,000) accounted for half of all visits to alternative practitioners.

The number of visits to any practitioner, per 1,000 population, increased from 2,373 in 1990 to 3,176 in 1997, an increase of 34%.  Overall, 46.3% of users of alternative medicine sought care from a practitioner in 1997, compared with 36.3% in 1990.

Specific medical conditions

In 1997, 42% of all alternative therapies used were attributed to treating specific medical problems, whereas 58% were used, at least in part, for prevention.  Overall, 31.8% of respondents who saw a medical doctor for a specific condition in 1997 also used an alternative therapy for that condition (up from 19.9% in 1990).  Of respondents who saw a practitioner of alternative therapy in 1997, 96% also saw a medical doctor. The majority of those who saw a medical doctor did not discuss their use of alternative therapy with this doctor.

Medical conditions for which alternative therapies were most likely to be used included:

  • Neck problems, reported by 12.1% of respondents in 1997. Of those who reported neck problems, 57% had used at least one alternative therapy during the previous 12 months, most commonly chiropractic and massage.
  • Back problems, reported by 24% in 1997. Of these, 47.6% had used an alternative therapy, again most commonly chiropractic and massage.
  • Anxiety (reported by 5.5%), depression (5.6%) and insomnia (9.3%). Resondents with these problems used alternative therapies quiet frequently -- 42.7% of those with anxiety, 40.9% of those with depression (both used mainly relaxation and spiritual healing) and 26.4% of those with insomnia (who used mainly relaxation and herbal medicine).
  • Other problems which were reported, and for which over 25% of those reporting the problem used alternative therapies in 1997 included fatigue (most commonly used therapies were relaxation and massage), arthritis (relaxation, chiropractic), headaches (relaxation, chiropractic) and digestive problems (relaxation, herbal medicine).

Cost considerations

In 1997, insurance coverage for alternative therapies was complete for 15.3%, partial for 26.4% and zero for 58.3%. 

Costs for visits to practitioners were estimated by two methods, a more conservative method and a less conservative one that was based on the RBRVS scale. According to the more conservative method, total expenditure on alternative medicine professional services in the United States was 21.2 billion dollars in 1997, compared with 14.6 billion dollars in 1990

Out-of-pocket expenses for alternative medicine include expenses for professional services not covered by insurance plus expenses for products such as vitamins, herbal medicines, commercial diets and books, classes and equipment. In 1997, using a conservative estimate, total out-of-pocket expenses for alternative medicine in the United States came to 27 billion dollars. This should be compared to estimated 1997 out-of-pocket expenses for all physician services of 29.3 billion dollars and out-of-pocket payments for hospitalization of 9.1 billion dollars.

Author's discussion

The authors note that both the prevalence and the expenditures for alternative medicine have increased substantially between 1990 and 1997.  The increase in cost is mainly the result of increased use of alternative techniques and increased use of the services of alternative  practitioners.

They note that the high prevalence of mega-vitamin and herbal remedy use often occurs in patients who are already taking prescription medication. Unfortunately, data on interactions between prescription medication and high-dose vitamins and herbs is scarce.

The high use of alternative medicine is occurring in the setting of low insurance coverage. Should these techniques be reimbursed to a greater extent, use is likely to increase even more.



This study, documenting the increasing degree to which Americans use alternative medicine, is the lead article in an issue of JAMA devoted entirely to alternative medicine.  This is clearly a hot topic, and articles on the subject are appearing in other journals as well. The NIH is also funding studies through the recently created Office of Alternative Medicine.  This study gives quantitative support to the impression that more and more patients are turning to non-traditional forms of health care.  Clearly, the attention being paid to alternative medicine is appropriate, at least in terms of the magnitude of its use.

In the same issue of JAMA is an article reporting on a randomized trial of moxibustion (burning herbs over certain acupuncture points) for correcting breech presentation. Burning herbs beside the outer corner of the fifth toenail was found to be efficacious in increasing fetal activity and promoting cephalic presentation.  I mention this only because it is an example of the unusual studies we are likely to be seeing in the near future.  Devising appropriate placebo controls for massage, relaxation and prayer will be challenging.

December 23, 1998 


References related to this article from the NLM's PubMed database. 

Reader Comments

Submit a comment

Journal Club homepage

Site Meter